How do thalamic neurons integrate their various feedforward and feedback inputs? Most of the inputs and synapses in the mammalian lateral geniculate nucleus (LGN) are extra-retinal, but the way in which these diverse inputs are integrated to control the flow of visual information from retina to cortex is not understood. In particular, the influence of the descending inputs from the cortex and the perigeniculate nucleus (PGN) on the spatiotemporal properties of receptive fields of LGN relay neurons is unknown, although the size and complexity of these inputs strongly suggest their importance, without them, the LGN might arguably be unnecessary. To address this knowledge gap, we shall combine physiological experiments with computational modeling to achieve the following aims: 1) Measure the spatiotemporal structure of receptive fields in the cat LGN before and during (reversible) inactivation of the descending feedback pathway from V1;2) Compare spatial summation in the retina with that of LGN neurons with and without V1 feedback;3) Measure the temporal transfer function of neurons in layer 6 of V1;4) Construct computational models of LGN relay neurons that incorporate the descending pathway from V1 and the PGN, and 5) Validate the models'predictions against physiological measurements. The proposed modeling, which builds on our initial (static) feedback model, will employ the innovative simulation approach of population kinetics, and will be one of the first attempts to model the corticothalamic feedback and its dynamics. It will use parallel computation on a scale not commonly found in neuroscience: two clusters of powerful computers, and a very large IBM supercomputer, which can accommodate larger, more complex models than could have been attempted in the past. The results will advance our understanding of the role that the descending inputs to the LGN play in establishing its sensitivity, dynamics, receptive field structure and discharge pattern, and will provide a necessary stepping stone for future expansions of our evolving model of the early visual system. HEALTH RELEVANCE: A more complete knowledge of how the LGN combines its diverse inputs, and especially how its temporal behavior depends on the cortex, should lead to insights into its role in dynamical brain diseases, such as epilepsy. More generally, an understanding of the role of feedback circuits will help us understand other dynamical pathologies, such as Parkinson's disease.